RT Conference Proceedings
T1 Towards a Smart Selection of Hybrid Platforms for Multimedia Processing
A1 Mahmoudi, Sidi Ahmed
A1 Manneback, Pierre
A2 Carretero Pérez, Jesús
A2 García Blas, Javier
A2 Petcu, Dana
AB Nowadays, images and videos have been present everywhere, they can come directly from camera, mobile devicesor from other peoples that share their images and videos. The latter are used to illustrate different objects in alarge number of situations. This makes from image and video processing algorithms a very important tool used forvarious domains related to computer vision such as video surveillance, medical imaging and database (images andvideos) indexation methods. The performance of these algorithms have been so reduced due the the high intensivecomputation required when using new image and video standards. In this paper, we propose a new framework thatallows users to select in a smart and efficient way the processing units (GPU or/and CPU) within heterogeneoussystems, when treating different kinds of multimedia objects : single image, multiple images, multiple videos andvideo in real time. The framework disposes of different image and video primitive functions that are implementedon GPU, such as shape (silhouette) detection, motion tracking using optical flow estimation, edges and cornersdetection. We have exploited these functions for several situations such as indexing videos, segmenting vertebraein in X-ray and MR images, detecting and localizing event in multi-user scenarios. Experimentation showedinteresting accelerations ranging from 6 to 118, by comparison with sequential implementations. Moreover, theparallel and heterogeneous implementations offered lower power consumption as a result for the fast treatment.
SN 978-84-608-6309-0
YR 2016
FD 2016-02
LK https://hdl.handle.net/10016/22881
UL https://hdl.handle.net/10016/22881
LA eng
NO Proceedings of the First PhD Symposium on Sustainable UltrascaleComputing Systems (NESUS PhD 2016) Timisoara, Romania. February 8-11, 2016.
NO European Cooperation in Science and Technology. COST
DS e-Archivo
RD 17 jul. 2024